End of training

README.md

@@ -5,6 +5,8 @@ license: apache-2.0
 metrics:
 - accuracy
 tags:
+- image-classification
+- vision
 - generated_from_trainer
 model-index:
 - name: only-lora-beans-vit-base-patch16-224-in21k
@@ -16,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # only-lora-beans-vit-base-patch16-224-in21k
 
-This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
+This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5903
-- Accuracy: 0.7895
+- Loss: 0.5416
+- Accuracy: 0.7744
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.7894736842105263,
-    "eval_loss": 0.5369813442230225,
-    "eval_runtime": 1.1887,
-    "eval_samples_per_second": 111.89,
-    "eval_steps_per_second": 14.302,
-    "total_flos": 8.772706474360013e+17,
-    "train_loss": 0.7462210376446063,
-    "train_runtime": 205.5107,
-    "train_samples_per_second": 50.314,
-    "train_steps_per_second": 6.326
+    "eval_accuracy": 0.7744360902255639,
+    "eval_loss": 0.5416202545166016,
+    "eval_runtime": 1.0948,
+    "eval_samples_per_second": 121.481,
+    "eval_steps_per_second": 15.528,
+    "total_flos": 8.400578669044531e+17,
+    "train_loss": 0.7472451694195087,
+    "train_runtime": 219.7142,
+    "train_samples_per_second": 47.061,
+    "train_steps_per_second": 5.917
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.7894736842105263,
-    "eval_loss": 0.5369813442230225,
-    "eval_runtime": 1.1887,
-    "eval_samples_per_second": 111.89,
-    "eval_steps_per_second": 14.302
+    "eval_accuracy": 0.7744360902255639,
+    "eval_loss": 0.5416202545166016,
+    "eval_runtime": 1.0948,
+    "eval_samples_per_second": 121.481,
+    "eval_steps_per_second": 15.528
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "total_flos": 8.772706474360013e+17,
-    "train_loss": 0.7462210376446063,
-    "train_runtime": 205.5107,
-    "train_samples_per_second": 50.314,
-    "train_steps_per_second": 6.326
+    "total_flos": 8.400578669044531e+17,
+    "train_loss": 0.7472451694195087,
+    "train_runtime": 219.7142,
+    "train_samples_per_second": 47.061,
+    "train_steps_per_second": 5.917
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.5369813442230225,
+  "best_metric": 0.5416202545166016,
   "best_model_checkpoint": "./beans_outputs/checkpoint-1170",
   "epoch": 10.0,
   "eval_steps": 500,
@@ -10,193 +10,193 @@
   "log_history": [
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 3.450658082962036,
+      "grad_norm": 3.9101290702819824,
       "learning_rate": 5e-05,
-      "loss": 1.0967,
+      "loss": 1.1012,
       "step": 100
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5413533834586466,
-      "eval_loss": 0.9973052740097046,
-      "eval_runtime": 0.9887,
-      "eval_samples_per_second": 134.519,
-      "eval_steps_per_second": 17.194,
+      "eval_accuracy": 0.5112781954887218,
+      "eval_loss": 1.0127754211425781,
+      "eval_runtime": 1.0349,
+      "eval_samples_per_second": 128.515,
+      "eval_steps_per_second": 16.427,
       "step": 130
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 1.2139726877212524,
+      "grad_norm": 1.3854057788848877,
       "learning_rate": 5e-05,
-      "loss": 1.0161,
+      "loss": 1.0265,
       "step": 200
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5789473684210527,
-      "eval_loss": 0.9221953749656677,
-      "eval_runtime": 1.1584,
-      "eval_samples_per_second": 114.811,
-      "eval_steps_per_second": 14.675,
+      "eval_accuracy": 0.6090225563909775,
+      "eval_loss": 0.8759572505950928,
+      "eval_runtime": 1.0786,
+      "eval_samples_per_second": 123.311,
+      "eval_steps_per_second": 15.762,
       "step": 260
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 1.6421459913253784,
+      "grad_norm": 2.5720674991607666,
       "learning_rate": 5e-05,
-      "loss": 0.9472,
+      "loss": 0.9263,
       "step": 300
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.6390977443609023,
-      "eval_loss": 0.8625620007514954,
-      "eval_runtime": 1.1687,
-      "eval_samples_per_second": 113.798,
-      "eval_steps_per_second": 14.546,
+      "eval_loss": 0.8891911506652832,
+      "eval_runtime": 1.1187,
+      "eval_samples_per_second": 118.885,
+      "eval_steps_per_second": 15.196,
       "step": 390
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 2.6952731609344482,
+      "grad_norm": 4.826686382293701,
       "learning_rate": 5e-05,
-      "loss": 0.8456,
+      "loss": 0.8131,
       "step": 400
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 1.875662088394165,
+      "grad_norm": 2.224466323852539,
       "learning_rate": 5e-05,
-      "loss": 0.7532,
+      "loss": 0.7348,
       "step": 500
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.7142857142857143,
-      "eval_loss": 0.7298586368560791,
-      "eval_runtime": 1.176,
-      "eval_samples_per_second": 113.091,
-      "eval_steps_per_second": 14.455,
+      "eval_accuracy": 0.706766917293233,
+      "eval_loss": 0.7197765111923218,
+      "eval_runtime": 1.2623,
+      "eval_samples_per_second": 105.362,
+      "eval_steps_per_second": 13.467,
       "step": 520
     },
     {
       "epoch": 4.615384615384615,
-      "grad_norm": 2.776125192642212,
+      "grad_norm": 3.2183878421783447,
       "learning_rate": 5e-05,
-      "loss": 0.6792,
+      "loss": 0.6856,
       "step": 600
     },
     {
       "epoch": 5.0,
       "eval_accuracy": 0.706766917293233,
-      "eval_loss": 0.6639184951782227,
-      "eval_runtime": 1.2266,
-      "eval_samples_per_second": 108.429,
-      "eval_steps_per_second": 13.859,
+      "eval_loss": 0.7358166575431824,
+      "eval_runtime": 1.192,
+      "eval_samples_per_second": 111.582,
+      "eval_steps_per_second": 14.262,
       "step": 650
     },
     {
       "epoch": 5.384615384615385,
-      "grad_norm": 2.474877119064331,
+      "grad_norm": 5.501096248626709,
       "learning_rate": 5e-05,
-      "loss": 0.6645,
+      "loss": 0.728,
       "step": 700
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.7218045112781954,
-      "eval_loss": 0.6237967610359192,
-      "eval_runtime": 1.15,
-      "eval_samples_per_second": 115.652,
-      "eval_steps_per_second": 14.783,
+      "eval_loss": 0.6856114268302917,
+      "eval_runtime": 1.0117,
+      "eval_samples_per_second": 131.468,
+      "eval_steps_per_second": 16.804,
       "step": 780
     },
     {
       "epoch": 6.153846153846154,
-      "grad_norm": 2.956679344177246,
+      "grad_norm": 3.860182523727417,
       "learning_rate": 5e-05,
-      "loss": 0.6473,
+      "loss": 0.6488,
       "step": 800
     },
     {
       "epoch": 6.923076923076923,
-      "grad_norm": 9.533896446228027,
+      "grad_norm": 7.866222858428955,
       "learning_rate": 5e-05,
-      "loss": 0.6056,
+      "loss": 0.6221,
       "step": 900
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.7443609022556391,
-      "eval_loss": 0.6903008818626404,
-      "eval_runtime": 1.1553,
-      "eval_samples_per_second": 115.117,
-      "eval_steps_per_second": 14.714,
+      "eval_accuracy": 0.6917293233082706,
+      "eval_loss": 0.8288138508796692,
+      "eval_runtime": 1.7977,
+      "eval_samples_per_second": 73.984,
+      "eval_steps_per_second": 9.457,
       "step": 910
     },
     {
       "epoch": 7.6923076923076925,
-      "grad_norm": 4.010281085968018,
+      "grad_norm": 3.9533588886260986,
       "learning_rate": 5e-05,
-      "loss": 0.614,
+      "loss": 0.6347,
       "step": 1000
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.706766917293233,
-      "eval_loss": 0.6278253793716431,
-      "eval_runtime": 1.1805,
-      "eval_samples_per_second": 112.66,
-      "eval_steps_per_second": 14.4,
+      "eval_accuracy": 0.7368421052631579,
+      "eval_loss": 0.5764672756195068,
+      "eval_runtime": 1.0499,
+      "eval_samples_per_second": 126.675,
+      "eval_steps_per_second": 16.192,
       "step": 1040
     },
     {
       "epoch": 8.461538461538462,
-      "grad_norm": 2.6526243686676025,
+      "grad_norm": 9.031426429748535,
       "learning_rate": 5e-05,
-      "loss": 0.6298,
+      "loss": 0.6217,
       "step": 1100
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.7894736842105263,
-      "eval_loss": 0.5369813442230225,
-      "eval_runtime": 0.995,
-      "eval_samples_per_second": 133.666,
-      "eval_steps_per_second": 17.085,
+      "eval_accuracy": 0.7744360902255639,
+      "eval_loss": 0.5416202545166016,
+      "eval_runtime": 1.0499,
+      "eval_samples_per_second": 126.682,
+      "eval_steps_per_second": 16.192,
       "step": 1170
     },
     {
       "epoch": 9.23076923076923,
-      "grad_norm": 4.809407711029053,
+      "grad_norm": 7.125640392303467,
       "learning_rate": 5e-05,
-      "loss": 0.6149,
+      "loss": 0.6099,
       "step": 1200
     },
     {
       "epoch": 10.0,
-      "grad_norm": 9.877020835876465,
+      "grad_norm": 15.404199600219727,
       "learning_rate": 5e-05,
-      "loss": 0.5867,
+      "loss": 0.5615,
       "step": 1300
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.7593984962406015,
-      "eval_loss": 0.579809308052063,
-      "eval_runtime": 1.1918,
-      "eval_samples_per_second": 111.6,
-      "eval_steps_per_second": 14.265,
+      "eval_accuracy": 0.7894736842105263,
+      "eval_loss": 0.5903372764587402,
+      "eval_runtime": 1.0014,
+      "eval_samples_per_second": 132.812,
+      "eval_steps_per_second": 16.976,
       "step": 1300
     },
     {
       "epoch": 10.0,
       "step": 1300,
-      "total_flos": 8.772706474360013e+17,
-      "train_loss": 0.7462210376446063,
-      "train_runtime": 205.5107,
-      "train_samples_per_second": 50.314,
-      "train_steps_per_second": 6.326
+      "total_flos": 8.400578669044531e+17,
+      "train_loss": 0.7472451694195087,
+      "train_runtime": 219.7142,
+      "train_samples_per_second": 47.061,
+      "train_steps_per_second": 5.917
     }
   ],
   "logging_steps": 100,
@@ -216,7 +216,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8.772706474360013e+17,
+  "total_flos": 8.400578669044531e+17,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null